The Multiple Pairwise Markov Chain Model Generalized Labeled Multi-Bernoulli Filter

Addressing the challenge of tracking multiple maneuvering targets with non-independent noise, an improved generalized labeled multi-Bernoulli (GLMB) filter, grounded in jump Markov system (JMS) and pairwise Markov chain (PMC) model, is developed in this paper. The proposed algorithm is composed of two parts. In the first part, the PMC model is introduced into GLMB filter to solve the non-independent noise problem in the tracking process by taking the joint variable containing the target state and the measurement information as a Markov process. In the second part, by modeling the motion process of targets as a system with multiple motion models, JMS is incorporated in the first part to tackle the problem of tracking multiple maneuvering targets with non-independent noise. The effectiveness of the proposed algorithm is demonstrated through simulation experiments.